Hi, everybody
OK, I got it working with "recursive". Don't know why this argument
slipped my mind, as I use filter() so often!
Now it is 44 times faster, which is good enough for me. :-)
Thank you, Gabor and Jim.
Best,
Sergey
On Fri, Jun 19, 2009 at 15:23, jim holtman wrote:
> check out 'filter
Dear Gabor and Jim
I am not looking at the "recursive" method for filter()
Recursive filter with lag 1 is specified in help files as:
y[i] = x[i] + f[1]*y[i-1]
My function looks like this:
EMA[i] = K*(C[i] - EMA[i-1]) + EMA[i-1],
that is:
y[i]=EMA[i]
y[i-1]=EMA[i-1]
x[i]=C[i]
So, I modified my
check out 'filter' to see if it does what you want with the 'recursive'
option.
On Fri, Jun 19, 2009 at 3:33 AM, Sergey Goriatchev wrote:
> Hello, everyone
>
> I have a long script that uses zoo objects. In this script I used
> simple moving averages and these I can very efficiently calculate wit
Hello, everyone
I have a long script that uses zoo objects. In this script I used
simple moving averages and these I can very efficiently calculate with
filter() functions.
Now, I have to use special "exponential" moving averages, and the only
way I could write the code was with a for-loop, which
4 matches
Mail list logo
